Learning curve and functional safety in robotic thyroidectomy via bilateral axillo-breast approach: a prospective study of 537 cases.

AIM: Robotic thyroidectomy via the bilateral axillo-breast approach (BABA) provides excellent cosmetic outcomes, but its adoption is limited by a steep learning curve and concerns about functional safety. This study aimed to define the learning curve trajectory and evaluate functional safety outcomes-with the primary endpoint of intraoperative RLN adverse events and secondary endpoints of inadvertent parathyroidectomy and PTH recovery.

METHODS

A prospective cohort study was conducted at a high-volume endocrine surgery center from March 2018 to March 2024. A total of 537 consecutive patients with differentiated thyroid carcinoma underwent robotic BABA thyroidectomy with central neck dissection. Intraoperative refinements included real-time recurrent laryngeal nerve (RLN) monitoring, intraoperative parathyroid-hormone (PTH) testing, and nanocarbon lymph node mapping. Learning phases were defined by cumulative sum analysis. Primary outcomes were intraoperative RLN adverse events and inadvertent parathyroidectomy; secondary outcomes included operative time, lymph node yield, PTH recovery, complication rates, and tracheal injury.

RESULTS

Among 537 patients (401 women, 74.7%; median age 43 years), operative time decreased from 189.9 to 129.3 min (P < 0.001), with plateaus at 152 and 352 cases. RLN adverse events declined from 16.7% (95% CI 7.9-30.2%) to 1.6% (95% CI 0.3-4.8%), and inadvertent parathyroidectomy decreased from 23.8% (95% CI 12.0-39.5%) to 1.6% (95% CI 0.3-4.8%). Six-month PTH levels improved from 3.15 to 4.14 pmol/L, and lymph node yield increased from 9.9 to 13.3 (P = 0.019). Three patients (0.6%) developed tracheal fistula.

CONCLUSIONS

Robotic BABA thyroidectomy demonstrates a structured learning curve in which functional safety improves with surgical experience and intraoperative standardization. These data-driven milestones may inform structured training and credentialing frameworks for robotic thyroid surgery.